Embedding Quality Research Articles

Knowledge graph representation learning with relation-guided aggregation and interaction

Knowledge graph representation learning (KGRL) aims to study the feature representations to solve the incompleteness of knowledge graphs (KGs). Recently, graph neural network (GNN) has achieved satisfactory performance in KGRL tasks. However, many GNN based KGRL approaches fail to capture the various semantic of entities, which may reduce the representation quality of embeddings. Furthermore, most of them treat relations equally when learning embeddings, which will lead to weakening of the specific semantic features. To alleviate the above problems, we propose a novel KGRL method (RGAI) with relation-guided aggregation (RGA) and relation-guided interaction (RGI). To capture rich attributes of entities, the RGA module generates multiple sub-structures according to relation types, and aggregates the semantic information among them. To mine specific semantic features and dig out the different contribution of relations, the RGI module calculates the importance of each relation to the central entity, and generates final entity features through aggregating representations of neighbors. Extensive experiments on standard KG datasets validate the effectiveness of the proposed innovations, and RGAI achieves state-of-the-art performance compared to existing approaches (e.g., compared to state-of-the-art approaches, RGAI improves Hits@10 from 0.548 to 0.562 on FB15K-237, Hits@3 from 0.561 to 0.582 on YAGO3-10, and Hits@1 from 0.449 to 0.455 on WN18RR).

0104 Biological Embedding and Sleep Quality in Adulthood: Evidence from a 30-year Prospective Longitudinal Cohort Study

0104 Biological Embedding and Sleep Quality in Adulthood: Evidence from a 30-year Prospective Longitudinal Cohort Study

GTLP: A graph transformer based link prediction framework with unified feature pre-processing

In the realm of graph representation learning, Graph Neural Networks (GNNs) have demonstrated exceptional efficacy across diverse tasks. Typically, GNNs employ message-passing schemes to disseminate node features along graph structures, culminating in learned graph representations. However, their heavy reliance on smoothed node features over graph structures, coupled with limited expressiveness in the presence of node attributes, often constrains link prediction performance. To surmount this challenge, we propose GTLP, a Graph Transformer based link prediction framework. GTLP integrates unsupervised GNNs and structure encoding, enabling a holistic consideration of both topological structures and node features. This approach preserves critical node location and role information, enhancing the model’s expressiveness. By introducing the Graph Transformer model, GTLP adeptly incorporates neighbor information, refining embedding quality and bolstering the model’s learning and generalization capabilities. Notably, our method exhibits superior scalability, accommodating diverse techniques for information extraction, embedding learning, and sampling. Experimental results underscore GTLP’s state-of-the-art performance, outpacing various baselines across five real-world datasets.

Hierarchical bottleneck for heterogeneous graph representation

Heterogeneous graphs (HGs) contain many nodes and their interaction relationships, which can model complex systems and provide rich semantic and structural information for task execution. Among these, HG representation stands as the fundamental and pivotal component. Existing HG representation methods primarily employ graph neural networks to acquire the semantics of nodes along various meta-paths and fuse them to represent the nodes. The most prevalent HG representation methods encompass two steps: semantic information extraction within meta-paths and semantic fusion between meta-paths. However, these methods overlooked the consideration of node heterogeneity within meta-paths and the simultaneous semantic correlation between meta-paths. Specifically, node heterogeneity within meta-paths signifies that the meta-path-based neighbors do not consistently contain information that positively influences the target node, and the semantic correlation between meta-paths indicates that different meta-path spaces are not entirely independent. Disregarding either of these issues leads to the propagation of irrelevant or redundant information and potential disruption of HG embedding. Consequently, in this study, we propose the HBHG, which is a hierarchical bottleneck for heterogeneous graph representation. HBHG primarily employs the information bottleneck (IB) as a guiding principle, constraining the propagation of irrelevant information within and between meta-paths while preserving relevant information. The central concept of the IB revolves around viewing model learning as the preservation of relevant information and compression of irrelevant information, accomplished by minimizing the dependency between input and hidden features through mutual information (MI) and maximizing the dependency between hidden features and ground-truth. Considering the complexity associated with MI estimation, this paper introduces a novel dependency index, namely the Hilbert-Schmidt independence criterion (HSIC), which offers ease of calculation. Specifically, HBHG comprises two primary components: a semantic bottleneck within meta-paths and a semantic bottleneck between meta-paths. The semantic bottleneck within meta-paths relies primarily on the HSIC-based limitations of dependencies at different layers of the graph neural network on various meta-paths, thereby maximizing the extraction of information relevant to the target node from neighboring nodes. The semantic bottleneck between meta-paths enables flexible extraction and fusion of semantic information based on downstream tasks, achieved by managing the trade-off of dependencies with HSIC between different meta-path semantic spaces. In summary, the proposed HBHG integrates hierarchical bottleneck constraints within and between meta-paths. This integration serves to maximize the aggregation of relevant information while effectively compressing irrelevant information, thereby enhancing the quality of heterogeneous graph embedding. The effectiveness of HBHG was validated through performance and ablation experiments conducted on multiple datasets.

Attributed Multi-Order Graph Convolutional Network for Heterogeneous Graphs

Heterogeneous graph neural networks play a crucial role in discovering discriminative node embeddings and relations from multi-relational networks. One of the key challenges in heterogeneous graph learning lies in designing learnable meta-paths, which significantly impact the quality of learned embeddings. In this paper, we propose an Attributed Multi-Order Graph Convolutional Network (AMOGCN), which automatically explores meta-paths that involve multi-hop neighbors by aggregating multi-order adjacency matrices. The proposed model first constructs different orders of adjacency matrices from manually designed node connections. Next, AMOGCN fuses these various orders of adjacency matrices to create an intact multi-order adjacency matrix. This process is supervised by the node semantic information, which is extracted from the node homophily evaluated by attributes. Eventually, we employ a one-layer simplifying graph convolutional network with the learned multi-order adjacency matrix, which is equivalent to the cross-hop node information propagation with multi-layer graph neural networks. Substantial experiments reveal that AMOGCN achieves superior semi-supervised classification performance compared with state-of-the-art competitors.

Multi-channel high-order network representation learning research.

The existing network representation learning algorithms mainly model the relationship between network nodes based on the structural features of the network, or use text features, hierarchical features and other external attributes to realize the network joint representation learning. Capturing global features of the network allows the obtained node vectors to retain more comprehensive feature information during training, thereby enhancing the quality of embeddings. In order to preserve the global structural features of the network in the training results, we employed a multi-channel learning approach to perform high-order feature modeling on the network. We proposed a novel algorithm for multi-channel high-order network representation learning, referred to as the Multi-Channel High-Order Network Representation (MHNR) algorithm. This algorithm initially constructs high-order network features from the original network structure, thereby transforming the single-channel network representation learning process into a multi-channel high-order network representation learning process. Then, for each single-channel network representation learning process, the novel graph assimilation mechanism is introduced in the algorithm, so as to realize the high-order network structure modeling mechanism in the single-channel network representation learning. Finally, the algorithm integrates the multi-channel and single-channel mechanism of high-order network structure joint modeling, realizing the efficient use of network structure features and sufficient modeling. Experimental results show that the node classification performance of the proposed MHNR algorithm reaches a good order on Citeseer, Cora, and DBLP data, and its node classification performance is better than that of the comparison algorithm used in this paper. In addition, when the vector length is optimized, the average classification accuracy of nodes of the proposed algorithm is up to 12.24% higher than that of the DeepWalk algorithm. Therefore, the node classification performance of the proposed algorithm can reach the current optimal order only based on the structural features of the network under the condition of no external feature supplementary modeling.

Assessing the Impact of Static, Contextual and Character Embeddings for Arabic Machine Translation

Word embeddings/representations are an important component of Natural Language Processing (NLP) tasks. Most Neural Machine Translation (NMT) systems that use such representations disregard word morphology by assigning a unique vector to each unique word in the used vocabulary and thus cannot handle Out-Of-Vocabulary (OOV) words. In some languages, such as Arabic, the meaning of words is associated with the meaning of the individual characters that constitute them, as these characters embody internal information. In this study, a combination of character- and word- level models is used to determine the most effective approaches to semantically and morphologically representing affective Arabic words. Furthermore, this work examines the strategy of combining static, character and contextual word embeddings to obtain richer representations for the Arabic Machine Translation (MT) task. To the best of our knowledge, we are the first to investigate the combination of static word embeddings, contextual ones and character-level representation in Arabic MT. Furthermore, a Deep Learning (DL) architecture is employed on data preprocessed by various prominent preprocessing techniques. Various experiments were conducted and the findings indicate that the integration of various models for word embedding and character-level representation is feasible and more effective than the state-of-the-art Arabic MT systems.

Artificial Intelligence and Machine Learning Methods to Evaluate Cardiotoxicity following the Adverse Outcome Pathway Frameworks.

Cardiovascular disease is a leading global cause of mortality. The potential cardiotoxic effects of chemicals from different classes, such as environmental contaminants, pesticides, and drugs can significantly contribute to effects on health. The same chemical can induce cardiotoxicity in different ways, following various Adverse Outcome Pathways (AOPs). In addition, the potential synergistic effects between chemicals further complicate the issue. In silico methods have become essential for tackling the problem from different perspectives, reducing the need for traditional in vivo testing, and saving valuable resources in terms of time and money. Artificial intelligence (AI) and machine learning (ML) are among today's advanced approaches for evaluating chemical hazards. They can serve, for instance, as a first-tier component of Integrated Approaches to Testing and Assessment (IATA). This study employed ML and AI to assess interactions between chemicals and specific biological targets within the AOP networks for cardiotoxicity, starting with molecular initiating events (MIEs) and progressing through key events (KEs). We explored methods to encode chemical information in a suitable way for ML and AI. We started with commonly used approaches in Quantitative Structure-Activity Relationship (QSAR) methods, such as molecular descriptors and different types of fingerprint. We then increased the complexity of encoders, incorporating graph-based methods, auto-encoders, and character embeddings employed in neural language processing. We also developed a multimodal neural network architecture, capable of considering the complementary nature of different chemical representations simultaneously. The potential of this approach, compared to more conventional architectures designed to handle a single encoder, becomes apparent when the amount of data increases.

NodeVector: A Novel Network Node Vectorization with Graph Analysis and Deep Learning

Network node embedding captures structural and relational information of nodes in the network and allows for us to use machine learning algorithms for various prediction tasks on network data that have an inherently complex and disordered structure. Network node embedding should preserve as much information as possible about important network properties where information is stored, such as network structure and node properties, while representing nodes as numerical vectors in a lower-dimensional space than the original higher dimensional space. Superior node embedding algorithms are a powerful tool for machine learning with effective and efficient node representation. Recent research in representation learning has led to significant advances in automating features through unsupervised learning, inspired by advances in natural language processing. Here, we seek to improve the representation quality of node embeddings with a new node vectorization technique that uses network analysis to overcome network-based information loss. In this study, we introduce the NodeVector algorithm, which combines network analysis and neural networks to transfer information from the target network to node embedding. As a proof of concept, our experiments performed on different categories of network datasets showed that our method achieves better results than its competitors for target networks. This is the first study to produce node representation by unsupervised learning using the combination of network analysis and neural networks to consider network data structure. Based on experimental results, the use of network analysis, complex initial node representation, balanced negative sampling, and neural networks has a positive effect on the representation quality of network node embedding.

A Biological Immunity-Based Neuro Prototype for Few-Shot Anomaly Detection with Character Embedding.

Anomaly detection has wide applications to help people recognize false, intrusion, flaw, equipment failure, etc. In most practical scenarios, the amount of the annotated data and the trusted labels is low, resulting in poor performance of the detection. In this paper, we focus on the anomaly detection for the text type data and propose a detection network based on biological immunity for few-shot detection, by imitating the working mechanism of the immune system of biological organisms. This network enabling the protected system to distinguish the aggressive behavior of "nonself" from the legitimate behavior of "self" by embedding characters. First, it constructs episodic task sets and extracts data representations at the character level. Then, in the pretraining phase, Word2Vec is used to embed the representations. In the meta-learning phase, a dynamic prototype containing encoder, routing, and relation is designed to identify the data traffic. Compare to the mean-based prototype, the proposed prototype applies a dynamic routing algorithm that assigns different weights to samples in the support set through multiple iterations to obtain a prototype that combines the distribution of samples. The proposed method is validated on 2 real traffic datasets. The experimental results indicate that (a) the proposed anomaly detection prototype outperforms state-of-the-art few-shot techniques with 1.3% to 4.48% accuracy and 0.18% to 4.55% recall; (b) under the premise of ensuring the accuracy and recall, the number of training samples is reduced to 5 or 10; (c) ablation experiments are designed for each module, and the results show that more accurate prototypes can be obtained by using the dynamic routing algorithm.

Context-Aware Attentive Multilevel Feature Fusion for Named Entity Recognition.

In the era of information explosion, named entity recognition (NER) has attracted widespread attention in the field of natural language processing, as it is fundamental to information extraction. Recently, methods of NER based on representation learning, e.g., character embedding and word embedding, have demonstrated promising recognition results. However, existing models only consider partial features derived from words or characters while failing to integrate semantic and syntactic information, e.g., capitalization, inter-word relations, keywords, and lexical phrases, from multilevel perspectives. Intuitively, multilevel features can be helpful when recognizing named entities from complex sentences. In this study, we propose a novel attentive multilevel feature fusion (AMFF) model for NER, which captures the multilevel features in the current context from various perspectives. It consists of four components to, respectively, capture the local character-level (CL), global character-level (CG), local word-level (WL), and global word-level (WG) features in the current context. In addition, we further define document-level features crafted from other sentences to enhance the representation learning of the current context. To this end, we introduce a novel context-aware attentive multilevel feature fusion (CAMFF) model based on AMFF, to fully leverage document-level features from all the previous inputs. The obtained multilevel features are then fused and fed into a bidirectional long short-term memory (BiLSTM)-conditional random field (CRF) network for the final sequence labeling. Extensive experiments on four benchmark datasets demonstrate that our proposed AMFF and CAMFF models outperform a set of state-of-the-art baseline methods and the features learned from multiple levels are complementary.

Hierarchically Contrastive Hard Sample Mining for Graph Self-Supervised Pretraining.

Contrastive learning has recently emerged as a powerful technique for graph self-supervised pretraining (GSP). By maximizing the mutual information (MI) between a positive sample pair, the network is forced to extract discriminative information from graphs to generate high-quality sample representations. However, we observe that, in the process of MI maximization (Infomax), the existing contrastive GSP algorithms suffer from at least one of the following problems: 1) treat all samples equally during optimization and 2) fall into a single contrasting pattern within the graph. Consequently, the vast number of well-categorized samples overwhelms the representation learning process, and limited information is accumulated, thus deteriorating the learning capability of the network. To solve these issues, in this article, by fusing the information from different views and conducting hard sample mining in a hierarchically contrastive manner, we propose a novel GSP algorithm called hierarchically contrastive hard sample mining (HCHSM). The hierarchical property of this algorithm is manifested in two aspects. First, according to the results of multilevel MI estimation in different views, the MI-based hard sample selection (MHSS) module keeps filtering the easy nodes and drives the network to focus more on hard nodes. Second, to collect more comprehensive information for hard sample learning, we introduce a hierarchically contrastive scheme to sequentially force the learned node representations to involve multilevel intrinsic graph features. In this way, as the contrastive granularity goes finer, the complementary information from different levels can be uniformly encoded to boost the discrimination of hard samples and enhance the quality of the learned graph embedding. Extensive experiments on seven benchmark datasets indicate that the HCHSM performs better than other competitors on node classification and node clustering tasks. The source code of HCHSM is available at https://github.com/WxTu/HCHSM.

Greedy routing optimisation in hyperbolic networks

Finding the optimal embedding of networks into low-dimensional hyperbolic spaces is a challenge that received considerable interest in recent years, with several different approaches proposed in the literature. In general, these methods take advantage of the exponentially growing volume of the hyperbolic space as a function of the radius from the origin, allowing a (roughly) uniform spatial distribution of the nodes even for scale-free small-world networks, where the connection probability between pairs decays with hyperbolic distance. One of the motivations behind hyperbolic embedding is that optimal placement of the nodes in a hyperbolic space is widely thought to enable efficient navigation on top of the network. According to that, one of the measures that can be used to quantify the quality of different embeddings is given by the fraction of successful greedy paths following a simple navigation protocol based on the hyperbolic coordinates. In the present work, we develop an optimisation scheme for this score in the native disk representation of the hyperbolic space. This optimisation algorithm can be either used as an embedding method alone, or it can be applied to improve this score for embeddings obtained from other methods. According to our tests on synthetic and real networks, the proposed optimisation can considerably enhance the success rate of greedy paths in several cases, improving the given embedding from the point of view of navigability.

When Protein Structure Embedding Meets Large Language Models.

Protein structure analysis is essential in various bioinformatics domains such as drug discovery, disease diagnosis, and evolutionary studies. Within structural biology, the classification of protein structures is pivotal, employing machine learning algorithms to categorize structures based on data from databases like the Protein Data Bank (PDB). To predict protein functions, embeddings based on protein sequences have been employed. Creating numerical embeddings that preserve vital information while considering protein structure and sequence presents several challenges. The existing literature lacks a comprehensive and effective approach that combines structural and sequence-based features to achieve efficient protein classification. While large language models (LLMs) have exhibited promising outcomes for protein function prediction, their focus primarily lies on protein sequences, disregarding the 3D structures of proteins. The quality of embeddings heavily relies on how well the geometry of the embedding space aligns with the underlying data structure, posing a critical research question. Traditionally, Euclidean space has served as a widely utilized framework for embeddings. In this study, we propose a novel method for designing numerical embeddings in Euclidean space for proteins by leveraging 3D structure information, specifically employing the concept of contact maps. These embeddings are synergistically combined with features extracted from LLMs and traditional feature engineering techniques to enhance the performance of embeddings in supervised protein analysis. Experimental results on benchmark datasets, including PDB Bind and STCRDAB, demonstrate the superior performance of the proposed method for protein function prediction.

Context-Focus Reframing for VUCA Era-Themed Texts on Discourse Analysis Assessment

Social values in the VUCA era can be adapted as self-understanding values to survive in various lines. Social categorization is a fundamental process that has important implications for various social phenomena. The role of social categories does not refer to a particular identity. Language learning that applies pragmatic aspects must accommodate assessments that refer to achieving the attitudes needed to face the VUCA era. This research will describe the discourse analysis assessment and observe its implications accurately. This research involved students from Indonesian, English, and Elementary School Teacher Education study programs in the Teacher Training and Education Faculty at Jember University. The results of this research illustrate the application of assessment aspects, namely determining relevant context, determining text implications by comparing social phenomena, conveying understood social values, and abstracting their components in a comment. This research shows that the role of assessment using context-focus reframing is very much needed in learning discourse analysis. Student answers indicate reasoning, accuracy of social values, accuracy of text structure, and argumentation through relevant statements as evaluation to support the comprehensive critical phase of the text after going through the literal and inference phases. In the next stage, perception scores regarding the VUCA era will be obtained based on the texts used as assessment material. This research shows students’ positive response to the assessment through context-focused reframing and character embedding through arguments to show attitudes through the question, “What are the implications for the VUCA era?”.

Rep-MCA-former: An efficient multi-scale convolution attention encoder for text-independent speaker verification

In many speaker verification tasks, the quality of speaker embedding is an important factor in affecting speaker verification systems. Advanced speaker embedding extraction networks aim to capture richer speaker features through the multi-branch network architecture. Recently, speaker verification systems based on transformer encoders have received much attention, and many satisfactory results have been achieved because transformer encoders can efficiently extract the global features of the speaker (e.g., MFA-Conformer). However, the large number of model parameters and computational latency are common problems faced by the above approaches, which make them difficult to apply to resource-constrained edge terminals. To address this issue, this paper proposes an effective, lightweight transformer model (MCA-former) with multi-scale convolutional self-attention (MCA), which can perform multi-scale modeling and channel modeling in the temporal direction of the input with low computational cost. In addition, in the inference phase of the model, we further develop a systematic re-parameterization method to convert the multi-branch network structure into the single-path topology, effectively improving the inference speed. We investigate the performance of the MCA-former for speaker verification under the VoxCeleb1 test set. The results show that the MCA-based transformer model is more advantageous in terms of the number of parameters and inference efficiency. By applying the re-parameterization, the inference speed of the model is increased by about 30%, and the memory consumption is significantly improved.

Robust Watermarking Algorithm for Building Information Modeling Based on Element Perturbation and Invisible Characters

With the increasing ease of building information modeling data usage, digital watermarking technology has become increasingly crucial for BIM data copyright protection. In response to the problem that existing robust watermarking methods mainly focus on BIM exchange formats and cannot adapt to BIM data, a novel watermarking algorithm specifically designed for BIM data, which combines element perturbation and invisible character embedding, is proposed. The proposed algorithm first calculates the centroid of the enclosing box to locate the elements, and establishes a synchronous relationship between the element coordinates and the watermarked bits using a mapping mechanism, by which the watermarking robustness is effectively enhanced. Taking into consideration both data availability and the need for watermark invisibility, the algorithm classifies the BIM elements based on their mobility, and perturbs the movable elements while embedding invisible characters within the attributes of the immovable elements. Then, the watermark information after dislocation is embedded into the data. We use building model and structural model BIM data to carry out the experiments, and the results demonstrate that the signal-to-noise ratio and peak signal-to-noise ratio before and after watermark embedding are both greater than 100 dB. In addition, the increased information redundancy accounts for less than 0.15% of the original data., which means watermark embedding has very little impact on the original data. Additionally, the NC coefficient of watermark extraction is higher than 0.85 when facing attacks such as translation, element addition, element deletion, and geometry–property separation. These findings indicate a high level of imperceptibility and robustness offered by the algorithm. In conclusion, the robust watermarking algorithm for BIM data fulfills the practical requirements and provides a feasible solution for protecting the copyright of BIM data.

On the use of big data frameworks in big service management

AbstractOver the last few years, big data have emerged as a paradigm for processing and analyzing a large volume of data. Coupled with other paradigms, such as cloud computing, service computing, and Internet of Things, big data processing takes advantage of the underlying cloud infrastructure, which allows hosting and managing massive amounts of data, while service computing allows to process and deliver various data sources as on‐demand services. This synergy between multiple paradigms has led to the emergence of big services, as a cross‐domain, large‐scale, and big data‐centric service model. Apart from the adaptation issues (e.g., need of high reaction to changes) inherited from other service models, the massiveness and heterogeneity of big services add a new factor of complexity to the way such a large‐scale service ecosystem is managed in case of execution deviations. Indeed, big services are often subject to frequent deviations at both the functional (e.g., service failure, QoS degradation, and IoT resource unavailability) and data (e.g., data source unavailability or access restrictions) levels. Handling these execution problems is beyond the capacity of traditional web/cloud service management tools, and the majority of big service approaches have targeted specific management operations, such as selection and composition. To maintain a moderate state and high quality of their cross‐domain execution, big services should be continuously monitored and managed in a scalable and autonomous way. To cope with the absence of self‐management frameworks for large‐scale services, the goal of this work is to design an autonomic management solution that takes the whole control of big services in an autonomous and distributed lifecycle process. We combine autonomic computing and big data processing paradigms to endow big services with self‐* and parallel processing capabilities. The proposed management framework takes advantage of the well‐known MapReduce programming model and Apache Spark and manages big service's related data using knowledge graph technology. We also define a scalable embedding model that allows processing and learning latent big service knowledge in a distributed manner. Finally, a cooperative decision mechanism is defined to trigger non‐conflicting management policies in response to the captured deviations of the running big service. Big services' management tasks (monitoring, embedding, and decision), as well as the core modules (autonomic managers' controller, embedding module, and coordinator), are implemented on top of Apache Spark as MapReduce jobs, while the processed data are represented as resilient distributed dataset (RDD) structures. To exploit the shared information exchanged between the workers and the master node (coordinator), and for further resolution of conflicts between management policies, we endowed the proposed framework with a lightweight communication mechanism that allows transferring useful knowledge between the running map‐reduce tasks and filtering inappropriate intermediate data (e.g., conflicting actions). The experimental results proved the increased quality of embeddings and the high performance of autonomic managers in a parallel and cooperative setting, thanks to the shared knowledge.

WSGMB: weight signed graph neural network for microbial biomarker identification.

The stability of the gut microenvironment is inextricably linked to human health, with the onset of many diseases accompanied by dysbiosis of the gut microbiota. It has been reported that there are differences in the microbial community composition between patients and healthy individuals, and many microbes are considered potential biomarkers. Accurately identifying these biomarkers can lead to more precise and reliable clinical decision-making. To improve the accuracy of microbial biomarker identification, this study introduces WSGMB, a computational framework that uses the relative abundance of microbial taxa and health status as inputs. This method has two main contributions: (1) viewing the microbial co-occurrence network as a weighted signed graph and applying graph convolutional neural network techniques for graph classification; (2) designing a new architecture to compute the role transitions of each microbial taxon between health and disease networks, thereby identifying disease-related microbial biomarkers. The weighted signed graph neural network enhances the quality of graph embeddings; quantifying the importance of microbes in different co-occurrence networks better identifies those microbes critical to health. Microbes are ranked according to their importance change scores, and when this score exceeds a set threshold, the microbe is considered a biomarker. This framework's identification performance is validated by comparing the biomarkers identified by WSGMB with actual microbial biomarkers associated with specific diseases from public literature databases. The study tests the proposed computational framework using actual microbial community data from colorectal cancer and Crohn's disease samples. It compares it with the most advanced microbial biomarker identification methods. The results show that the WSGMB method outperforms similar approaches in the accuracy of microbial biomarker identification.

Enhancing efficiency and capacity of telehealth services with intelligent triage: a bidirectional LSTM neural network model employing character embedding

BackgroundThe widespread adoption of telehealth services necessitates accurate online department selection based on patient medical records, a task requiring significant medical knowledge. Incorrect triage results in considerable time wastage for both patients and medical professionals. To address this, we propose an intelligent triage model based on a Bidirectional Long Short-Term Memory (Bi-LSTM) neural network with character embedding to enhance the efficiency and capacity of telehealth services.MethodsWe gathered a 1.3 GB medical dataset comprising 200,000 records, each including medical history, physical examination data, and other pertinent information found on the electronic medical record homepage. Following data preprocessing, a clinical corpus was established to train character embeddings with a medical context.These character embeddings were then utilized to extract features from patient chief complaints, and a 2-layer Bi-LSTM neural network was trained to categorize these complaints, enabling intelligent triage for telehealth services.Results60,000 chief complaint-department data pairs were extracted from clinical corpus and divided into the training, validation, and test sets of 42,000, 9,000, and 9,000, respectively. The character embedding based Bi-LSTM neural network achieved a macro-precision of 85.50% and an F1 score of 85.45%. ConclusionThe telehealth triage model developed in this study demonstrates strong implementation outcomes and significantly improves the efficiency and capacity of telehealth services. Character embedding outperforms word embedding, and future work will incorporate additional features such as patient age and gender into the chief complaint feature to future enhance model performance.